From eaa74ec7329a48a4b724d8de440b3a2cbaabf7c8 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Thu, 21 Jul 2016 13:03:09 -0700 Subject: [PATCH 1/5] IB/core: Make rdma_rw_ctx_init() initialize all used fields Some but not all callers of rdma_rw_ctx_init() zero-initialize struct rdma_rw_ctx. Hence make rdma_rw_ctx_init() initialize all work request fields that will be read by ib_post_send(). Fixes: a060b5629ab0 ("IB/core: generic RDMA READ/WRITE API") Signed-off-by: Bart Van Assche Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Tested-by: Steve Wise Tested-by: Laurence Oberman Cc: Parav Pandit Cc: Nicholas Bellinger Cc: #v4.7+ Signed-off-by: Doug Ledford --- drivers/infiniband/core/rw.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/core/rw.c b/drivers/infiniband/core/rw.c index 1eb9b1294a63..1ad2baaa6c8c 100644 --- a/drivers/infiniband/core/rw.c +++ b/drivers/infiniband/core/rw.c @@ -71,6 +71,7 @@ static inline u32 rdma_rw_fr_page_list_len(struct ib_device *dev) return min_t(u32, dev->attrs.max_fast_reg_page_list_len, 256); } +/* Caller must have zero-initialized *reg. */ static int rdma_rw_init_one_mr(struct ib_qp *qp, u8 port_num, struct rdma_rw_reg_ctx *reg, struct scatterlist *sg, u32 sg_cnt, u32 offset) @@ -114,6 +115,7 @@ static int rdma_rw_init_mr_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, struct scatterlist *sg, u32 sg_cnt, u32 offset, u64 remote_addr, u32 rkey, enum dma_data_direction dir) { + struct rdma_rw_reg_ctx *prev = NULL; u32 pages_per_mr = rdma_rw_fr_page_list_len(qp->pd->device); int i, j, ret = 0, count = 0; @@ -125,7 +127,6 @@ static int rdma_rw_init_mr_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp, } for (i = 0; i < ctx->nr_ops; i++) { - struct rdma_rw_reg_ctx *prev = i ? &ctx->reg[i - 1] : NULL; struct rdma_rw_reg_ctx *reg = &ctx->reg[i]; u32 nents = min(sg_cnt, pages_per_mr); @@ -162,9 +163,13 @@ static int rdma_rw_init_mr_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp, sg_cnt -= nents; for (j = 0; j < nents; j++) sg = sg_next(sg); + prev = reg; offset = 0; } + if (prev) + prev->wr.wr.next = NULL; + ctx->type = RDMA_RW_MR; return count; @@ -205,11 +210,10 @@ static int rdma_rw_init_map_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp, rdma_wr->wr.opcode = IB_WR_RDMA_READ; rdma_wr->remote_addr = remote_addr + total_len; rdma_wr->rkey = rkey; + rdma_wr->wr.num_sge = nr_sge; rdma_wr->wr.sg_list = sge; for (j = 0; j < nr_sge; j++, sg = sg_next(sg)) { - rdma_wr->wr.num_sge++; - sge->addr = ib_sg_dma_address(dev, sg) + offset; sge->length = ib_sg_dma_len(dev, sg) - offset; sge->lkey = qp->pd->local_dma_lkey; @@ -220,8 +224,8 @@ static int rdma_rw_init_map_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp, offset = 0; } - if (i + 1 < ctx->nr_ops) - rdma_wr->wr.next = &ctx->map.wrs[i + 1].wr; + rdma_wr->wr.next = i + 1 < ctx->nr_ops ? + &ctx->map.wrs[i + 1].wr : NULL; } ctx->type = RDMA_RW_MULTI_WR; From 632bc3f65081dd1e2e5394a9161580a0f78e8839 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Thu, 21 Jul 2016 13:03:30 -0700 Subject: [PATCH 2/5] IB/core, RDMA RW API: Do not exceed QP SGE send limit Compute the SGE limit for RDMA READ and WRITE requests in ib_create_qp(). Use that limit in the RDMA RW API implementation. Signed-off-by: Bart Van Assche Cc: Christoph Hellwig Cc: Sagi Grimberg Cc: Steve Wise Cc: Parav Pandit Cc: Nicholas Bellinger Cc: Laurence Oberman Cc: #v4.7+ Reviewed-by: Christoph Hellwig Signed-off-by: Doug Ledford --- drivers/infiniband/core/rw.c | 10 ++-------- drivers/infiniband/core/verbs.c | 9 +++++++++ include/rdma/ib_verbs.h | 6 ++++++ 3 files changed, 17 insertions(+), 8 deletions(-) diff --git a/drivers/infiniband/core/rw.c b/drivers/infiniband/core/rw.c index 1ad2baaa6c8c..dbfd854c32c9 100644 --- a/drivers/infiniband/core/rw.c +++ b/drivers/infiniband/core/rw.c @@ -58,13 +58,6 @@ static inline bool rdma_rw_io_needs_mr(struct ib_device *dev, u8 port_num, return false; } -static inline u32 rdma_rw_max_sge(struct ib_device *dev, - enum dma_data_direction dir) -{ - return dir == DMA_TO_DEVICE ? - dev->attrs.max_sge : dev->attrs.max_sge_rd; -} - static inline u32 rdma_rw_fr_page_list_len(struct ib_device *dev) { /* arbitrary limit to avoid allocating gigantic resources */ @@ -186,7 +179,8 @@ static int rdma_rw_init_map_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u64 remote_addr, u32 rkey, enum dma_data_direction dir) { struct ib_device *dev = qp->pd->device; - u32 max_sge = rdma_rw_max_sge(dev, dir); + u32 max_sge = dir == DMA_TO_DEVICE ? qp->max_write_sge : + qp->max_read_sge; struct ib_sge *sge; u32 total_len = 0, i, j; diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 6298f54b4137..e39a0b597234 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -814,6 +814,15 @@ struct ib_qp *ib_create_qp(struct ib_pd *pd, } } + /* + * Note: all hw drivers guarantee that max_send_sge is lower than + * the device RDMA WRITE SGE limit but not all hw drivers ensure that + * max_send_sge <= max_sge_rd. + */ + qp->max_write_sge = qp_init_attr->cap.max_send_sge; + qp->max_read_sge = min_t(u32, qp_init_attr->cap.max_send_sge, + device->attrs.max_sge_rd); + return qp; } EXPORT_SYMBOL(ib_create_qp); diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 7e440d41487a..e694f02d42e3 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1428,6 +1428,10 @@ struct ib_srq { } ext; }; +/* + * @max_write_sge: Maximum SGE elements per RDMA WRITE request. + * @max_read_sge: Maximum SGE elements per RDMA READ request. + */ struct ib_qp { struct ib_device *device; struct ib_pd *pd; @@ -1449,6 +1453,8 @@ struct ib_qp { void (*event_handler)(struct ib_event *, void *); void *qp_context; u32 qp_num; + u32 max_write_sge; + u32 max_read_sge; enum ib_qp_type qp_type; }; From 30c6d8773de06878f920666d8c945f81cb2081b3 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Thu, 21 Jul 2016 13:03:47 -0700 Subject: [PATCH 3/5] IB/srpt: Limit the number of SG elements per work request Limit the number of SG elements per work request to what the HCA and the queue pair support. Fixes: 34693573fde0 ("IB/srpt: Reduce QP buffer size") Reported-by: Parav Pandit Signed-off-by: Bart Van Assche Cc: Christoph Hellwig Cc: Sagi Grimberg Cc: Steve Wise Cc: Parav Pandit Cc: Nicholas Bellinger Cc: Laurence Oberman Cc: #v4.7+ Reviewed-by: Christoph Hellwig Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/srpt/ib_srpt.c | 3 ++- drivers/infiniband/ulp/srpt/ib_srpt.h | 6 +++++- 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index 4a4155640d51..9a3b954e862d 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -1601,6 +1601,7 @@ static int srpt_create_ch_ib(struct srpt_rdma_ch *ch) struct ib_qp_init_attr *qp_init; struct srpt_port *sport = ch->sport; struct srpt_device *sdev = sport->sdev; + const struct ib_device_attr *attrs = &sdev->device->attrs; u32 srp_sq_size = sport->port_attrib.srp_sq_size; int ret; @@ -1638,7 +1639,7 @@ static int srpt_create_ch_ib(struct srpt_rdma_ch *ch) */ qp_init->cap.max_send_wr = srp_sq_size / 2; qp_init->cap.max_rdma_ctxs = srp_sq_size / 2; - qp_init->cap.max_send_sge = SRPT_DEF_SG_PER_WQE; + qp_init->cap.max_send_sge = min(attrs->max_sge, SRPT_MAX_SG_PER_WQE); qp_init->port_num = ch->sport->port; ch->qp = ib_create_qp(sdev->pd, qp_init); diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.h b/drivers/infiniband/ulp/srpt/ib_srpt.h index 389030487da7..581878782854 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.h +++ b/drivers/infiniband/ulp/srpt/ib_srpt.h @@ -106,7 +106,11 @@ enum { SRP_LOGIN_RSP_MULTICHAN_MAINTAINED = 0x2, SRPT_DEF_SG_TABLESIZE = 128, - SRPT_DEF_SG_PER_WQE = 16, + /* + * An experimentally determined value that avoids that QP creation + * fails due to "swiotlb buffer is full" on systems using the swiotlb. + */ + SRPT_MAX_SG_PER_WQE = 16, MIN_SRPT_SQ_SIZE = 16, DEF_SRPT_SQ_SIZE = 4096, From 10fce586b20ba27013f0a6ac73b3ac87c95cf8b5 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Thu, 21 Jul 2016 13:04:06 -0700 Subject: [PATCH 4/5] IB/srpt: Simplify srpt_queue_response() Initialize first_wr to &send_wr. This allows to remove a ternary operator and an else branch. This patch does not change the behavior of srpt_queue_response(). Signed-off-by: Bart Van Assche Reviewed-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Tested-by: Steve Wise Tested-by: Laurence Oberman Cc: Parav Pandit Cc: Nicholas Bellinger Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/srpt/ib_srpt.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index 9a3b954e862d..dfa23b075a88 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -2262,7 +2262,7 @@ static void srpt_queue_response(struct se_cmd *cmd) container_of(cmd, struct srpt_send_ioctx, cmd); struct srpt_rdma_ch *ch = ioctx->ch; struct srpt_device *sdev = ch->sport->sdev; - struct ib_send_wr send_wr, *first_wr = NULL, *bad_wr; + struct ib_send_wr send_wr, *first_wr = &send_wr, *bad_wr; struct ib_sge sge; enum srpt_command_state state; unsigned long flags; @@ -2303,11 +2303,8 @@ static void srpt_queue_response(struct se_cmd *cmd) struct srpt_rw_ctx *ctx = &ioctx->rw_ctxs[i]; first_wr = rdma_rw_ctx_wrs(&ctx->rw, ch->qp, - ch->sport->port, NULL, - first_wr ? first_wr : &send_wr); + ch->sport->port, NULL, first_wr); } - } else { - first_wr = &send_wr; } if (state != SRPT_STATE_MGMT) From e6d66e3eb65f8b083d827f6864e70b8dcea9d9bb Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Thu, 21 Jul 2016 13:04:25 -0700 Subject: [PATCH 5/5] IB/isert: Remove an unused member variable Signed-off-by: Bart Van Assche Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Tested-by: Steve Wise Tested-by: Laurence Oberman Cc: Parav Pandit Cc: Nicholas Bellinger Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/isert/ib_isert.c | 2 -- drivers/infiniband/ulp/isert/ib_isert.h | 1 - 2 files changed, 3 deletions(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index a990c04208c9..ba6be060a476 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -137,8 +137,6 @@ isert_create_qp(struct isert_conn *isert_conn, attr.cap.max_recv_wr = ISERT_QP_MAX_RECV_DTOS + 1; attr.cap.max_rdma_ctxs = ISCSI_DEF_XMIT_CMDS_MAX; attr.cap.max_send_sge = device->ib_device->attrs.max_sge; - isert_conn->max_sge = min(device->ib_device->attrs.max_sge, - device->ib_device->attrs.max_sge_rd); attr.cap.max_recv_sge = 1; attr.sq_sig_type = IB_SIGNAL_REQ_WR; attr.qp_type = IB_QPT_RC; diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h index e512ba941f2f..fc791efe3a10 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.h +++ b/drivers/infiniband/ulp/isert/ib_isert.h @@ -138,7 +138,6 @@ struct isert_conn { u32 responder_resources; u32 initiator_depth; bool pi_support; - u32 max_sge; struct iser_rx_desc *login_req_buf; char *login_rsp_buf; u64 login_req_dma;